发现大模型时间序列预测易受对抗攻击，且扰动极小却破坏力强。
Adversarial Vulnerabilities in Large Language Models for Time Series Forecasting
- 设计黑箱无梯度攻击框架，生成微小扰动破坏预测
- 多模型多数据集测试显示性能下降远超随机噪声
- 揭示大模型在时序预测中的安全漏洞，适合关注AI可信性的研究者
大型语言模型（LLMs）在时间序列预测中展现出巨大潜力，能有效处理复杂时序数据。然而，其在真实场景下的鲁棒性和可靠性仍缺乏深入研究，尤其对对抗攻击的敏感性尚未明确。本文提出一种针对基于LLM的时间序列预测的定向对抗攻击框架，采用无梯度与黑盒优化方法，生成极小但高度有效的扰动，显著降低多个数据集和不同LLM架构（如LLMTime、GPT-3.5、GPT-4、LLaMa、Mistral、TimeGPT、TimeLLM）的预测准确率。实验表明，对抗攻击导致的性能退化远超随机噪声，且在多种模型上均具广泛有效性。结果凸显了LLMs在时间序列预测中的关键脆弱性，强调需建立可靠防御机制以保障实际部署的安全性。代码仓库见：https://github.com/JohnsonJiang1996/AdvAttack_LLM4TS。
原文摘要 · Abstract (English)
Large Language Models (LLMs) have recently demonstrated significant potential in time series forecasting, offering impressive capabilities in handling complex temporal data. However, their robustness and reliability in real-world applications remain under-explored, particularly concerning their susceptibility to adversarial attacks. In this paper, we introduce a targeted adversarial attack framework for LLM-based time series forecasting. By employing both gradient-free and black-box optimization methods, we generate minimal yet highly effective perturbations that significantly degrade the forecasting accuracy across multiple datasets and LLM architectures. Our experiments, which include models like LLMTime with GPT-3.5, GPT-4, LLaMa, and Mistral, TimeGPT, and TimeLLM show that adversarial attacks lead to much more severe performance degradation than random noise, and demonstrate the broad effectiveness of our attacks across different LLMs. The results underscore the critical vulnerabilities of LLMs in time series forecasting, highlighting the need for robust defense mechanisms to ensure their reliable deployment in practical applications. The code repository can be found at https://github.com/JohnsonJiang1996/AdvAttack_LLM4TS.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。